The overall goal of this proposal is to understand the respective contributions of aging, chronic HIV infection, and antiretroviral therapy to both skeletal and non-skeletal risk for fracture among older men. Our previous studies have shown that osteoporosis is 3 to 4 times more common in HIV-infected persons compared to those without HIV, and that the prevalence of fragility fractures in HIV-infected persons is about 70% greater than in an HIV-uninfected population. Many previous studies regarding fracture risk in HIV-infected patients have relied on dual- energy x-ray absorptiometry (DXA) to evaluate bone mineral density (BMD). Although clinically useful, BMD accounts for only about 50% of fracture risk. Other important skeletal characteristics, including bone strength and turnover as well as non-skeletal characteristics, such as the risk of falling have not been adequlately evaulated in older HIV-infected populations to date. In addition, the mechanisms underlying both skeletal and non-skeletal risk for fracture among HIV-infected persons, such as alterations in body composition and inflammation/immune activation, require futher investigation. To address these unresolved issues, we propose a substudy of older men with or at-risk for HIV-infection, nested within the Multicenter AIDS Cohort Study with these specific aims: (1)To determine whether skeletal risk factors for fracture, including a) bone density (measured by both DXA and quantitative computerized tomography (QCT)), b) bone strength (measured by specialized analysis of hip DXA and QCT), and c) bone turnover differ by HIV status and to determine whether age modifies differences by HIV status;(2) To determine whether non-skeletal risk factors for fracture, including balance, strength, and lower extremity performance differ by HIV status, and whether age modifies differences by HIV status;(3) To determine whether differences in body composition between HIV-infected and -uninfected men mediate the differences by HIV-status in skeletal and non-skeletal risk factors for fracture;(4) To determine whether systemic inflammation and immune activation/senescence are associated with and are predictive of skeletal and non-skeletal risk factors for fracture in older men and whether these relationships differ by HIV-status. Using a well-characterized cohort with an internal HIV-uninfected reference population, the proposed studies will improve our understanding of fracture risk in HIV infection, leading to effective strategies for fracture prevention.